The ethics of complexity and the complexity of ethics | Minka Woermann & Paul Cilliers (2012)

Reproduced from:

Cite as: Woermann, M., & Cilliers, P. (2012). The ethics of complexity and the complexity of ethics. South African Journal of Philosophy, 31(2), 447-463. 

The ethics of complexity and the complexity of ethics

Minka Woermann

Department of Philosophy
Stellenbosch University
Private Bag X1
Matieland 7602
Stellenbosch; South Africa


Paul Cilliers

Department of Philosophy
Stellenbosch University
Post-mortem publication


In this paper, we investigate the implications that a general view of complexity – i.e. the view that complex phenomena are irreducible – hold for our understanding of ethics. In this view, ethics should be conceived of as constitutive of knowledge and identity, rather than as a normative system that dictates right action. Using this understanding, we elaborate on the ethics of complexity and the complexity of ethics. Whilst the former concerns the nature and the status of our modelling choices, the latter denotes a contingent and recursive understanding of ethics. Although the complexity of ethics cannot be captured in a substantive normative model, we argue that this view of ethics nevertheless commits one to, what we term, ‘the provisional imperative’. Like Kant’s categorical imperative, the provisional imperative is substantively-empty; however, unlike Kant’s imperative, our imperative cannot be used to generate universal ethical principles. As such, the provisional imperative simultaneously demands that we must be guided by it, whilst drawing attention to the exclusionary nature of all imperatives. We further argue that the provisional imperative urges us to adopt a certain attitude with regard to ethical decision-making, and that this attitude is supported and nurtured by provisionality, transgressivity, irony, and imagination.


Ethics, complexity, provisional imperative, provisionality, transgressivity, irony, imagination, trust


The term ‘complexity’ is often loosely appropriated to describe things that lack simple explanations. More specifically, the paradigm of complexity offers a challenge to traditional reductive explanations, which are premised on the assumption that complex systems can be completely understood in terms of their component parts. If we cannot know phenomena in their full complexity, then an engagement with complexity thinking implies a critical engagement with the status and limits of our knowledge claims. However, the challenge posed by complexity thinking moves beyond this general epistemological level, to influence the whole system of thought that defines our specific thinking on matters related to our practices, politics, and ethics (Morin, 2008).

It is particularly this latter issue – i.e. ethics – that is of interest in the context of this paper. When we utilise a complexity perspective in our thinking about the world, we are busy with a task that is both descriptive and normative in nature. As soon as we engage with complexity, we have to make certain modelling choices when describing phenomena. In other words, since we cannot have complete knowledge of complex things, we cannot “calculate” their behaviour in any deterministic fashion. We have to interpret and evaluate. Our decisions always involve an element of choice that cannot be justified objectively, but are, in part, based on normative judgements. Otherwise stated, our modelling choices are based on subjective judgements about what matters – both in terms of our work and in terms of our personal lives. This introduces an unavoidable ethical component into our thinking about complex phenomena (Preiser & Cilliers, 2010; Cilliers, de Villiers & Roodt, 2002; Derrida, 1988). In this regard, ethics should be understood as something that constitutes both our knowledge and us, rather than as a normative system that dictates right action. Hence, the ethics of complexity is not an add-on, but inherent to any real engagement with complex phenomena. Otherwise stated, the ethics of complexity is a structural element of complexity thinking. In practice, this means that we should assume a critical attitude when modelling phenomena, where the critical attitude amounts to both the recognition of, and engagement with, the limits of knowledge (Preiser & Cilliers, 2010).

In this paper, we investigate, and offer ways in which to deal with the challenges that arise from an engagement with the ethics of complexity. The most important implication in this regard concerns the possibility of a substantive ethics. In simple terms, this amounts to the fact that, although complexity thinking necessarily involves ethics, it cannot provide any information regarding the content of such an ethics since our sense of right and wrong, good and bad, and what deserves respect and what does not, cannot be justified on a priori grounds. Moreover, such a position implies that any substantive notion of ethics must itself be subjected to a deconstruction of sorts, since our ethical models are limited and, hence, exclusionary (Cilliers, 2005; Derrida, 2005; 2002a; 2002b; 1999). As such the logic which informs the ethics of complexity commits us to accepting the complexity of ethics. Although our position prevents us from giving a substantive account of ethics, we argue that the critical position that we develop nevertheless constitutes a type of ethical strategy, similar to Immanuel Kant’s (1993) categorical imperative, which urges us to adopt a certain attitude when undertaking ethical decisions. However, unlike Kant, we argue for the provisionality of the ethical imperative, and further show that such an imperative is served by the mechanisms of provisionality, transgressivity, irony, and the imagination (Preiser & Cilliers, 2010). Although these mechanisms can help us to remain sensitive to the complexities that define the contexts in which we operate, there can be no guarantee that ethical actions will ensue. On the final count, the complexity of ethics (which is so intimately interlinked with the rich diversity of what it means to be human) can only be fostered through nurturing trusting relations and through an active recognition of, and engagement with, difference. This paper therefore concludes with a discussion on the importance of trust as a virtue to be cultivated within a complex world.

Understanding Complexity and The Importance of the Critical Enterprise

In order to better understand the claim that the ethics of complexity is a structural element of complexity thinking, it is useful to follow Edgar Morin (2007) in distinguishing between two perspectives on complexity, namely restricted complexity and general complexity. The central difference between these two paradigms concerns how we view the status of our practices.

According to Morin (2007), the goal of a restricted approach to complexity is to study the multiple, interrelated processes that constitute complex systems, in order to retroactively uncover the rules or laws of complexity. This approach is popular amongst the researchers at the Sana Fe Institute (which was founded in 1984, and which is dedicated to the study of complex systems), and much of the work conducted at the Institute is dedicated to discovering, comprehending and communicating the common fundamental principles in a variety of systems, which underlie many of the pressing problems currently facing science and society.

Whilst research conducted at the Institute has no doubt led to important advances in formalisation and modelling, Jack Cowan (in Horgan, 1995: 104), one of the Institute’s founders, notes that the major discovery to have emerged from the Institute is the insight that ‘it’s very hard to do science on complex systems’, if by science one understands the process of discovering and modelling the rules and laws that govern the behaviour of all phenomena. Such a view is informed by, what John Horgan (1995) calls, a seductive syllogism, which is based on the premises that since a computer that follows a simple set of mathematical rules can give rise to extremely complicated patterns, and since extremely complicated patterns exhibit in the world, simple rule must also give rise to complicated worldly phenomena. In their implicit acceptance of this syllogism, many researchers at the Institute, thus adhere to a restricted approach to complexity, in that it is believed that, with enough time and effort, we will be able to construct a unified theory of complexity – also referred to as the ‘Theory of Complexity’ (TOC) or the ‘Theory of Everything’ (TOE) (Chu, Strand & Fjelland, 2003). In other words, the hope is that complex phenomena can be encapsulated in a precise definition or mathematical equation.

We support a notion of general complexity and argue that it is impossible to construe a strict science of complex systems, if by science one understands the practice of uncovering the rules and laws that govern all phenomena. Although we cannot conclusively state that complexity is an ontological category of the world as opposed to merely a consequence of our epistemological limitations, this does not imply that we can relegate complexity to the status of a mere practical problem (which can ultimately be solved with enough computing power). One cannot simply ‘cut-up’ complex systems in order to understand them, since what is of interest is the dynamic, local interrelations that exist between the parts of a complex system, and which give rise to emergent phenomena (which are often not reducible to base laws). In this process, contingency – expressed in terms of both intra-systemic and extra-systemic conditions (Wimsatt, 2008) – also plays a crucial role, which further frustrates any efforts to merely calculate the resultant effect of a certain configuration of parts. Therefore, in terms of general complexity, any attempt at formulating a TOC will necessarily fail because complexity itself is not accounted for (Morin, 2007).

Complex phenomena are irreducible, or, to elaborate in the words of the theoretical biologist, Robert Rosen (1985: 424), a system is complex precisely ‘to the extent that it admits non-equivalent encodings; encodings which cannot be reduced to one another.’ Moreover, we argue that this view commits one to a self-critical rationality, which is defined as a rationality that makes no claim for objectivity, or for any special status for the grounds from which the claim is made. A self-critical rationality is therefore the outcome of acknowledging the irreducible nature of complexity.

The fact that an engagement with complexity is not a purely objective exercise, does not however imply an ‘anything goes’ approach. Indeed, as Peter Allen (2000: 93) states, ‘[a] representation or model with no assumptions whatsoever is clearly simply subjective reality’ and therefore ‘does not concern systemic knowledge’. Acknowledging the irreducible nature of complexity means that we should ‘apply our “complexity reduction” assumptions honestly’ (94), rather than accept the defeatist attitude that limited knowledge commits us to relativism. In this context, intellectual honesty implies modesty, which denotes sensitivity to the levels and limits of our understanding. In other words, we must still be competent at performing the necessary calculations and considering the relevant information, but we should also recognise that doing the groundwork won’t resolve the complexity and that we still remain responsible for our modelling choices, since each choice gives rise to ‘a different spectrum of possible consequences, different successes and failures, and different strengths and weaknesses’ (102). Knowledge acquisition is not the objective pursuit of truth, but rather a process of working towards finding suitable strategies for dealing with complex phenomena. However, since there is no final model, and since knowledge is a tumultuous building site, Morin (2007:21) argues that we must introduce a double conscience into our practices: ‘a conscience of itself and an ethical conscience.’

The ethics of complexity is an expression of this double conscience, since it implies an acknowledgement of the implications that a general view of complexity holds for both the status of our models and for our attitude towards these models, and compels us to remain perpetually vigilant in the face of uncertainty. Vigilance demands a continual and critical evaluation and transformation of our claims and practices, and, in this context, David Wood’s (1999: 117) description of a Derridean notion of responsibility applies to our understanding of the ethics of complexity, which ‘is not quantifiably (or even inquantifiably) large and, therefore, not a basis of guilt through failure to live up to it. It is rather a recursive modality, an always renewable openness’. This renewable openness is safeguarded by a self-critical rationality, and the ethics of complexity therefore commits us to not just a general understanding, but also a critical understanding of complexity.

The (Im)Possibility of Ethics

The previous section can be summarised as follows: a critical notion of complexity implies a radical or recursive understanding of ethics, which amounts to the insight that we cannot do away with ethical considerations because we can neither compute knowledge nor appeal to a priori principles to justify our knowledge claims. Every claim implies a choice. Furthermore, since our models must necessarily exclude some of the complexity, we are responsible for our choices, and we exercise this responsibility through critique, and moreover, self-critique. However, the ethics of complexity cannot do more than generate awareness of the fact that we are always in trouble. In other words, it cannot provide substantive guidelines for an ethical system. The question that now arises is whether we can move beyond this position, in order to say something more about the complexity of ethics. We argue that, despite not being able to provide a substantive ethics, it is possible to develop a type of meta-ethical position, which serves to highlight important considerations that underscore the ethical strategies that we employ when engaging in the particularities of situations.

As a starting point, we turn to perhaps the most famous example of a meta-ethical position in the history of moral philosophy, namely Kant’s (1993) categorical imperative. The categorical imperative is a substantively-empty rule, in that it cannot generate contingent ethical principles, but can merely act as a yardstick for evaluating the morality of principles which already exist. This is because Kant wants his moral rule to be categorically applicable and, hence, universally valid. However, the only rule which conforms to this criterion is a purely abstract and formal rule, which says ‘always follow only universal rules’ or, otherwise stated, ‘always follow only rules which you will want all other people to follow’. Thus, by combining a purely formal rule with the notion of universability, Kant can generate a formulation that actually does say something about ethics, namely that if certain contingent principles are universalisable, then the principles are deemed morally correct. Therefore, although the categorical imperative cannot indicate which principles are good, right, and deserving of respect, it does provide a strategy for evaluating our contingent principles. As such, one can argue that Kant’s categorical imperative urges us to adopt a certain strategy when undertaking moral considerations (Preiser & Cilliers, 2010).

Next, we can try to apply the same logic Kant uses to the ethics of complexity, in order to say something about the complexity of ethics (in other words, in order to develop a meta-ethical position). From the analysis thus far we can construct the following argument: all knowledge (including self-knowledge) is limited because, in order to generate meaning, we need to reduce the complexity through modelling. Our models are radically contingent in time and space because they are the product of the resources at our disposal, the choices that we make, and the influences that act upon us (including the influences of others). Since all knowledge is contingent, it is also subject to revision, and therefore irreducibly provisional. Following Kantian logic, we can now capture the gist of the above argument in the following imperative: ‘When acting, always remain cognisant of other ways of acting’. Our meta-ethical position thus constitutes a provisional imperative (Preiser & Cilliers, 2010).

Note that on one reading, the idea of a provisional imperative is a contradiction in terms, since the logic of an imperative is absolute: either you follow the imperative or you don’t. The idea of a provisional imperative seems to suggest that the imperative itself is subject to change, and in this regard we seem to be advocating an impossible position. This is, to a large extent, exactly the point: we cannot do away with moral imperatives, but, if we take complexity seriously, we should also realise that our imperatives are the outcome of certain framing strategies or ways of thinking about the world, and are thus necessarily exclusionary. Thus the provisional imperative stipulates that we must be guided by the imperative, whilst simultaneously acknowledging the exclusionary nature of all imperatives.

In terms of the actual content of the imperative, it must be noted that – unlike the Kantian imperative – which tells us something about the rules for action, the provisional imperative says something about our state of mind or attitude when choosing rules for action. Again: it is impossible to say that ‘When acting, always choose rules that admit to the possibility of other rules’, since the logic of rules (as with the logic of imperatives) is absolute. In this regard, Jacque Derrida (1988: 116) notes that:

Every concept that lays claim to any rigor whatsoever implies the alternative of “all or nothing”… Even the concept of “difference to degree,” the concept of relativity is, qua concept, determined according to the logic of all or nothing, of yes or no: differences of degree or no differences of degree. It is impossible or illegitimate to form a philosophical concept outside the logic of all or nothing.

In the above citation, Derrida is pointing to structural conditions of all concepts. We cannot do other than model and exclude. Yet, what the provisional imperative tells us is that when we act, we must be cognisant of this logic.

In this regard, we argue that it makes a difference – and moreover, an ethical difference – whether one exercises this awareness. This is because if we remain open to other ways of modelling and other ways of being, we are more likely to practice a self-critical rationality, to respect diversity, to be willing to revise our models, and to guard against the naturalisation of these models. The provisional imperative, therefore, provides us with a strategy for remaining open to complexity at the same time that we reduce complexity through our decisions and actions.

Four Operations in Service of the Critical Position

Practically, it is very demanding to commit to a recursive or deconstructive view of ethics. Acknowledging that we are always in trouble can be daunting; and, in this regard, it is helpful to take note of four mechanisms that serve to reinforce and promote the critical attitude, namely provisionality, transgressivity, irony, and imagination.


Although the recursive or deconstructive view of ethics developed above necessarily implies provisionality, it is helpful to unpack the meaning of provisionality in more detail, in order to shed further light on the nature of our imperative. As stated above, provisionality is the outcome of the contingent nature of our knowledge claims. The source of this contingency is two-fold: firstly, the meaning of our claims is dependent on the context in which they are made. Language is iterable, which means that although concepts are repeatable and understandable across contexts, the meaning of the concept shifts every time the concepts are used; or, in the words of Mikhail Bakhtin (1984: 202):

The life of the word is contained in its transfer… from one context to another context… In this process the word does not forget its own path and cannot completely free itself from the power of these concrete contexts into which it has entered.

This, as Derrida (1979: 81) notes, implies that ‘no meaning can be determined out of context’. A good illustration of this concerns how we understand the term ‘freedom’, used as the title of Jonathan Franzen’s (2010) recent book, compared to its use in the title of Mandela’s (1994) autobiography, ‘A Long Walk to Freedom’. In both counts, the term ‘freedom’ is familiar, but in the former use it denotes a critical appraisal of contemporary American society; whereas in the latter use, the term is associated with the liberation struggle, and the story of Mandela’s own imprisonment.

Secondly – and more radically – meaning is contingent because even within a given context we cannot fix meaning, since, as Derrida (1979:81) argues, ‘no context permits saturation’. Due to the complexities involved, every context is open to further description, and meanings change as the interpretation of the context changes. With regard to our example, one can argue that our understanding of ‘freedom’ in Mandela’s autobiography is dependent on our own personal background. Not only does the understanding of the concept vary from one person to the next, but the very same book can never be reread in exactly the same way: what Mandela’s bibliography, and the importance of freedom meant to me ten years ago, will differ from what it will mean if I were to reread the book today. Provisionality therefore draws attention to both the spatial and the temporal dimensions of meaning (and by extension, of ethics).

What the above implies is that ‘[t]here are no final meanings that arrest the movement of signification’ (Culler, 1983: 188). There is no universal origin that we can somehow access through applying ourselves. Unlike Plato (1987), who argues that the ideal form of the Good is the ultimate object of knowledge, a complex understanding of ethics posits the good as something which is necessarily subject to revision and deconstruction; and, hence as something which is provisional. Although pockets of consensus and relative stability are needed for the designation of right and wrong, it is important not to naturalise our ethical positions or arrest the movement of signification in the name of Truth. This is especially significant given the nature of today’s geopolitics in which Western ideals all-too-often pass as universal ideals. In this regard, it is useful to recall Derrida’s (2002c: 10) view of philosophy as something which is ‘no more assigned to its origin or by its origin, than it is simply, spontaneously, or abstractly cosmopolitical or universal.’ He continues in arguing that:

There are other ways for philosophy than those of appropriation or expropriation… Not only are there other ways for philosophy, but philosophy, if there is any such thing, is the other way.

In our context, we can substitute the term ‘philosophy’ with ‘ethics’, since what lies at the heart of the provisional imperative is the belief that ethics is indeed the other way; or, more poignantly, the way which is still to come.


Preiser and Cilliers (2010) write that the critical position informed by complexity will have to be transgressive. It can never simply re-enforce that which is current, but – as the definition states – involves a violation of accepted or imposed boundaries. In this regard, transgressivity demands bold action. It can be argued that, on a literal level, transgressivity is at odds with modesty, which – as discussed earlier – also underscores the critical position. However, in order to practice transgressivity responsibly, one must be modest enough to recognise the limitations of one’s conceptual schema, and show a willingness to overcome these limitations. Modesty and transgressivity thus go hand-in-hand, since modesty acts as the impetus for transgressivity, in focusing attention on the possibility of other rules of action (as commanded by the provisional imperative).

Moreover, being transgressive is not only an ethical move, but also a political move. It involves recognition of the importance of remaining vigilant and open to diversity and to the future, whilst simultaneously exercising choice and taking in a position. Derrida (2002d: 29) calls this the aporia of politics and ethics, which exhibits in the deconstructive nature of negotiation, defined as ‘a work of mediation… a to-and-fro between impatience and patience’. Transgressivity demands absolute engagement with both ethics and politics, since both are concerned with the here and the now, and require a thoughtful and urgent answer to the question ‘What should I do?’ (Derrida, 2002b: 296; 302). Transgressivity, like deconstruction, is therefore situated in the fold of the aporia between affirmation (ethics) and position (politics) (Derrida, 2002d: 25). Furthermore, transgressivity is what gives ethics potence, since, as Derrida argues (25), ‘[a]ffirmation requires a position. It requires that one move to action and do something, even if it is imperfect’. We need to act, even though we know we cannot get it right (Preiser & Cilliers, 2010: 270). The double-logic of ethics and politics thus marks the heart of the critical position.

Alain Badiou (2009) also hints at the importance of transgressivity in his moving tribute to Derrida. In this tribute, he describes Derrida as the opposite of a hunter, because unlike the hunter – who hopes that the animal will arrest its movements so that it can be shot – Derrida’s animal cannot cease fleeing. This is because locating the animal does not mean grasping; rather, it means suppressing it. This metaphor is intended to explain Derrida’s passion for doing justice to what Badiou terms the ‘non-existent’. Framing or modelling phenomena necessarily implies a reduction of complexity. In making choices, we leave out certain considerations from our models and, in the social realm, these considerations may include the interests of stakeholders, who – in terms of our model – become the non-existent or those who are not accounted for in our conceptual paradigms. However, the moment we try to do justice to the non-existent by accounting for their interests, we assimilate the alterity of the Other into our system of thought. In other words, the outside becomes inside. In terms of Badiou’s metaphor, the animal is the Other – to whom we must do justice, even though we cannot know the Other in her alterity. The endless flight, or ethics as a recursive modality, is therefore served by transgressivity.

Furthermore, non-existence sits within the aporia of ethics and politics. On the one hand, to say that the ‘the non-existent is’ fails to convey that it does not exist – at least not in terms of any explicit conceptual paradigm; and, on the other hand, to say that ‘the non-existent does not exist’ fails to convey that it is (141) (i.e. that despite not being heard, the interests of unseen stakeholders, for example, continue to exist). Non-existence therefore simultaneously necessitates vigilance of the fact that not all the complexities of a situation are accounted for, and proactively attempts to give voice to the disenfranchised (which, for Badiou (141), is typified in the war cry of the Revolution, which reads ‘We are nothing, let us be all!’). Transgressivity, as an attempt to do homage to non-existence, is therefore fed by the complexity of ethics, but leads to the binaries of politics.

Lastly, being transgressive implies that we acknowledge the fact that there is no way in which we can fully engage with the excess of meaning that results from complexity; that our context (i.e. our position) is defined by certain freedom and constraints, which acts as the necessary conditions for action and transformation; and, that we have to acknowledge and exercise choice. These three acknowledgements are interdependent: complexity is not the result of indeterminacy. Rather, as Derrida (1999: 79) states, the problem is that ‘there is too much determinacy.’ There is no mystical basis for complexity. Complexity is only the result of the complex interrelations between the components of the system, and of our attempts to model this complexity (which, as Luhmann (2000) notes, creates further complexities due to the fact that our models are imperfect renditions of complex systems). As a result, Derrida (1999: 79) writes that ‘there are many meanings struggling with one another, there are tensions, there are overdeterminations, there are equivocations’. Paradoxically, it is these overdeterminations that generate freedom, since exercising choice and assuming a position is possible because of the constraints within which we operate. In this regard, Morin (2008: 113) writes that ‘the complex notion of self-organization permits us to conceive of beings that are relatively autonomous as beings while remaining subject to the necessities and hazards of existence’. Complexity is generated by both dependence and freedom, and in order to understand the nature of choice, these concepts must be thought simultaneously.

Transgressivity does not imply an abandonment of everything that came before. Rather, in order to responsibly transgress current systems of meaning, we need to concede to the inextricable ways in which our lives are constituted by the systems of meaning in which we partake, whilst nevertheless remaining vigilant of the fact that we have both the duty to continually break open and transform these systems in order to account for the non-existent, as well as the duty to take responsibility for our positions – even when they have undesirable and unforeseen consequences. As Derrida (1981: 12) explains:

There is not a transgression, if one understands by that a pure and simple landing into a beyond of metaphysics… Now, even in aggressions or transgressions, we are consorting with a code to which metaphysics is tied irreducibly, such that every transgressive gesture reencloses us – precisely by giving us a hold on the closure of metaphysics – within this closure. But, by means of the work done on one side and the other of the limit the field inside is modified and a transgression is produced that consequently is nowhere present as a fait accompli. One is never installed within transgression, one never lives elsewhere. Transgression implies that the limit is always at work.


Transgressivity is supported by an ironic outlook on life. Consider dictionary descriptions of irony as a means of expressing something other than the literal intention of words, as a demonstration of incongruence between what is expected and what is, and as an expression of that which is contrary to plan or expectation. In these descriptions, irony is defined as a means by which to subvert the idea of an objective reality. This is achieved by introducing an element of contingency and play into literal, objective language. In formulating the provisional imperative, we stated that the logic of a rule is absolute: either you follow the rule or not. Yet, the value of irony is that it draws attention to the supplementary complications that govern all rules and that point to the impossibility of concluding any general theory that rules give rise to (Derrida, 1988). As such, irony affirms the necessity of improvising when faced with the limits of a binary logic.

It is in this regard that irony can be interpreted as a generative creative task, akin to a form of improvisation, and one can further argue that there is ‘an important and potentially fruitful connection’ between these skills and ‘the lived experience of complexity’ (Montuori, 2003: 238). Alfonso Montuori (238) elaborates in saying that ‘improvisation and creativity are capacities we would do well to develop in an increa ingly unpredictable, complex, and at times chaotic experience’. It is specifically in relation to developing fruitful and responsible strategies for living that the ironic dimension of the critical task is indispensible.

The theme of irony is not new to philosophy. Indeed, one of the earliest uses of irony is demonstrated in Plato’s dialogues, where the Socrates’ rhetorical technique ‘was [used] to pretend ignorance and, more sneakily, to feign credence in your opponent’s power of thought, in order to tie him in knots’ (Williams, 2003). Furthermore, the German Philosopher, Friedrich Schlegel, popularised the notion of romantic or philosophical irony, which he employed as a ‘more complex philosophical tool’, used to shed light on the divided self and a multiplicity of perspectives that could potentially unlock the truth of the whole (Williams, 2003). The concept of irony is also prevalent in the work of Hegel, Kierkegaard, Nietzsche, and – more recently – Richard Rorty, the latter arguing that ‘irony is the only possible ethic of modern liberalism’ (Colebrook, 2004: 151). Despite all these well-known accounts of irony in the extant literature, we turn to a figure that is not generally recognised as an ironist in the philosophical literature, namely Claude Levi-Strauss’s bricoleur, in order to further our analysis.

A bricoleur is defined as a fiddler or tinker, and, by extension, as someone who makes creative and resourceful use of whatever materials are at hand. Levi-Straus (1966) distinguishes the bricoleur from the engineer, arguing that the former approximates the savage mind that operates in a closed universe and is therefore forced to do with the means at hand, whereas the latter approximates the scientific mind that operates in an open universe because he can develop new tools, and thereby construct the totality of the universe. In ‘Structure, sign, and play’, Derrida (1978) deconstructs the distinction between the engineer and the bricoleur, by arguing that because there is no absolute origin (i.e. no objective reality that we can access through our practices), the engineer remains a theological idea; or, more radically, ‘the odds are that the engineer is a myth produced by the bricoleur’ (285). Derrida’s point is that the world is a complex place, and pretending that it is otherwise is also merely a bricolage or strategy for dealing with the world. As such, Derrida (285) writes that:

as soon as we admit that every finite discourse is bound by a certain bricolage and that the engineer and the scientist are also species of bricoleurs, then the very idea of bricolage is menaced and the difference in which it took on its meaning breaks down.

There can be no a priori basis from which to argue for the merits of one life strategy over another. However, it is important to assume responsibility for, and bear the consequences of, our choices and decisions. This is only possible if we are aware of the nature and status of our strategies, which Morin (2008: 96) refers to as ‘the art of working with uncertainty’. Irony is therefore a way of moving beyond a binary logic and of expressing the double movement involved in affirming a certain position or life strategy, whilst simultaneously undermining the absolutist status of that which we affirm through our lives. We are all improvisers who not only tell a story, but become a story. We create interwoven narratives, which together, constitute a tapestry of stories (Montuori, 2003; also see Kearney, 1988). The ethical moment lies in whether we concede to this or not, i.e. whether we accept – with irony and humour – our limited knowledge and fragile personal experiences, and focus these in the very moment that we are living in (Montuori, 2003). To be able to improvise and to live with irony:

requires a different discipline, a different way of organizing our thoughts and actions. It requires, and at best elicits, a social virtuosity which reflects our state of mind, our perceptions of who we are, and a willingness to take risks, to let go of the safety of the ready-made, the already written, and to think, create, and ‘write’ on the spot (244).

Assuming an ironic disposition simultaneously draws attention to the status of our strategies, and lightens the burden of self-awareness. This is because those who live with irony find it easier to confess to the fact that their lives are not following a determinate course, but represent the outcome of choices and decisions. Irony – like transgressivity – needs modesty, which in this context means adopting a self-deprecating humour, and not taking oneself or one’s ideas too seriously, as this may prevent one from exercising the openness and tolerance1 needed to act responsibly in the face of complexity. Irony is a critical task, without which we potentially open the door to human evil. This is because, as Susan Sontag (2007: 227) suggests, it is exactly ‘this refusal of an extended awareness’ (which she defines as taking in ‘more than is happening right now, right here’) that lies at the heart of ‘our ever-confused awareness of evil’ and ‘of the immense capacity of human beings to commit evil’. In a sense then, it is irony that allows us to face up to the seriousness of our responsibilities, which is an insight which accords beautifully with the description of irony as a demonstration of incongruence between what is expected and what is.


Whereas irony assists us in forging strategies for successfully engaging within the constraints that characterise the present, imagination allows us to undertake the creative leap necessary to engage with a future that we cannot calculate. No one can contest the urgent need to move towards a more sustainable future, and in this regard, the role of the imaginative dimension of the critical task is crucial. This is because, as Allen (2000: 103) notes, ‘creativity’ – and we argue, specifically imagination – ‘is the motor of change, and the hidden dynamic that underlies the rise and fall of civilizations, peoples, and regions, and evolution both encourages and feeds on invention’.

In this respect, imagination constitutes the ability to generate variety and options, and to break out of one’s ‘closed or limited hermeneutic circles’ (Verstraeten, 2000: ix). Indeed, without imagination it is impossible to practice the provisional imperative, which commands us to take cognisance of other rules and other ways of being. In this regard, it is useful to distinguish between ‘requisite diversity’, which denotes the minimal level of variety needed for a system to cope with its environment, and ‘excess diversity’, which allows systems to experiment internally and thereby generate a number of strategies for operating in a given environment. Allen (2001) argues that excess diversity is needed for long-term systems survival, since the ‘fat’ of excess knowledge and diversity is needed both for breaking out of our conceptual schema and for imagining, and thereby experimenting and innovating for the future.

Todd May (1995: 145) notes that ‘[t]he terms in which one thinks of oneself and one’s possibilities, the practical parameters of those possibilities, and the ease or difficulty in realizing them, are all social as well as individual matters’. Therefore, imagination – defined as the generation of excess variety – is both a psychological and a social process. On the psychological level, we can develop our imagination by engaging in the arts, which – far from being a pleasurable diversion – is an important way in which to break out of our hermeneutic circles, or – otherwise stated – to ‘transform the framework we apply when apprehending the world’ (Cilliers, 2005: 264). This is because, as Morin (1999) notes, the creative arts foster awareness of human complexity, and draw attention to the full range of human subjectivity. As an example, Morin argues that fictional criminals – such as the gangster kings of Shakespeare, the royal gangsters of films noirs, Jean Valjean and Raskolnikov – are portrayed in all their fullness in literature and film, rather than the least or worst part of themselves (as is often the case with real life criminals). Morin also uses the example of the movie tramp, Charlie Chaplin, in order to illustrate how films use psychological techniques of projection and identification, which brings us to understand and sympathise with people that we would normally find foreign or disgusting. As such, books and films help us ‘to learn the greatest lesson of life: compassion and true understanding for the humiliated in their suffering’ (53). To this we add that the value of the arts lie not only in understanding human complexity, but also the complexities that define our situatedness in the world. As an example, consider Edward Burtynsky’s (2006) photos of manufactured landscapes, which document humanity’s impact on the world, and which persuaded millions to join a global conversation on sustainability. In this regard, imagination can help us to foster not only social sustainability, but also to think about what environmental sustainability might mean. Imagination – like irony – is therefore not just about seeing other ways of being, but also about creating new ways of being.

On the social level, we follow Timothy Hargrave (2009: 87) in defining moral imagination as a process, which ‘emerges through dialectical processes that are influenced by actors’ relative power and political skill’. Again, moral imagination is a skill that needs to be fostered and exercised within ‘pluralistic processes in which multiple actors with opposing moral viewpoints interact, and [where] no single actor is in control’ (90). An element of conflict is also always present in imaginative activities because of the ‘lived tensions between contradictory perspectives’ (91). Although Hargrave views moral imagination in terms of a collective action model, his analysis of moral imagination also has implications for individuals. In this regard, Hargrave (91) argues that ‘morally imaginative actors recognise and integrate contradictory moral viewpoints, and also integrate moral sensitivity… [of] contextual considerations’. Since these characteristics are also hallmarks of critical thought, one can argue that moral imagination is itself a critical activity (Woermann, 2010). Another characteristic of moral imagination is that it involves an element of uncertainty or risk. Far from being a form of creative abandonment, moral imagination necessitates that we critically project and plan for the future (Woermann, 2010). However, since this future cannot be known, and since uncertainty involves a real property of situations, we have to respond with judgement (Luntley, 2003).

Since the future is uncertain we should be tolerant of each other’s opinions, and also tolerant of failure. Edmund Husserl (in Mensch, 2003: 143) explains that tolerance means that I affirm ‘his ideals as his, as ideals which I must affirm in him, just as he must affirm my ideals – not, indeed, as his ideals of life but as the ideals of my being and life.’ James Mensch (2003) explains that, in Latin, tolerance has the sense of supporting or sustaining, rather than enduring or suffering. He further states that ‘it can be understood as the attitude that actively sustains the maximum number of compatible possibilities of being human’ (142). As such, tolerance should be understood as the ideal of human fullness. The reason why Mensch views human fullness as an ideal is because it demands more than can be achieved by a single individual (for example, I cannot simultaneously realise the possibility of being a professional weightlifter and a sprinter). Therefore, for Mensch, tolerance ‘appears when we acknowledge our finitude in attempting to embody this ideal’ (142-143), as well as when we recognise the uniqueness and singularity of human beings.

Tolerance is thus essential in allowing the personal and social imagination to flourish, since without it, it is impossible to generate the excess variety needed to productively engage in one’s environment. From this argument, we can also deduce that tolerance is the acceptance of human complexity, even though this complexity can never be fully understood, but only imagined. In this regard, tolerance is nurtured by the creative arts and flourishes in diverse human societies, in which freedom and other aspects of disorder are accepted, and in which innovation and creativity blossom in defiance of perspectives that try to frame societies as fixed, homogenous systems (Morin, 2008). Imagination and tolerance is therefore that which safeguards us from ‘the well-managed dystopia of the brave new world’ (Cilliers, 2005: 264).

Practicing the Provisional Imperative

The above analysis demonstrates how provisionality, transgressivity, irony, and imagination can help us to remain cognisant of other ways of acting (and hence, help us to remain faithful to the demands of the provisional imperative). In this last section, the importance of the provisional imperative for our relations with others is investigated. We begin by referring to an anecdote cited by the cybernetican, Heinz von Foerster (1990):

I have a friend who grew up in Marrakech. The house of his family stood on the street that divide [sic] the Jewish and Arabic quarter. As a boy he played with all the others, listened to what they thought and said, and learned of their fundamentally different views. When I asked him once, “Who was right” he said, “They are both right.”

“But this cannot be,” I argued from an Aristotelian platform, “Only one of them can have the truth!”

“The problem is not truth,” he answered, “The problem is trust.”

This anecdote points both to the limitations of a binary logic, and the need to overcome these limitations, in order to foster trust. The emphasis on trust, as opposed to truth is an interesting choice, and if we look at the basic characteristics of trust, we see why it is an important virtue for governing relations in a complex world. In her celebrated paper, entitled ‘Trust and Antitrust’, Annette Baier (1986) argues that trust creates vulnerability, but that this vulnerability is inevitable, because, following two simple Socratic truths, we need the help of others in creating, and caring for the things that we value, and, therefore, have no choice but to place ourselves in a position where others can harm us. Furthermore, because of the richness of human diversity, it follows that not everyone places value on the same things. In complexity terms, one can say that there are many different ways in which to model our world, and these models are interrelated.

This also holds important implications for understanding identity. Our identities are neither a priori nor static. Rather, identity is constituted in a complex network, and must be contextualised as both a temporal process of becoming, and as a point in a nexus of relationships (Cilliers, 2010). We act on one another in ways that give rise to our personal identities as well as the identities of our social practices, and that leads to the transformation of these identities (Woermann, 2010). We are therefore not only vulnerable with regard to the things that we value, but also with regard to our very identities. My state depends on the state of others (Preiser & Cilliers, 2010). Here, one must draw a distinction between objects of entrustments and trust relations. Whereas the former implies a measure of discretionary responsibility (B knowing what is entrusted (C) by A), and, in some cases, a form of relative power (where A is dependent in some sense upon B’s goodwill) (Baier, 1986); the latter is characterised by tolerance (that B will act in such a way so as to allow A to flourish). Whereas, entrustment implies responsibility for something, trust relations imply a responsibility towards someone (Painter-Morland, 2006). We can only exercise our responsibility toward others if we show a fundamental respect for difference, even when our actions reduce difference. The only way in which to exercise this responsibility is to heed to the provisional imperative, and to always remain cognisant of other ways of acting.

Of course not all differences are good, and in this regard, the provisional imperative – like the Kantian imperative – can be used as a yardstick to evaluate our and other people’s choices and actions. Since the provisional imperative is substantively empty, it should be supplemented by concrete ethical positions that can inform our notions of concrete morality. However, when these ethical positions lead to the homogenization of ‘the different perspectives until everybody thinks, believes, and acts the same way’ (von Foerster & Poerksen, 2002: 36), they will necessarily fail the demands imposed on us by the provisional imperative and should therefore be morally condemned. Indeed, Derrida (1988: 119) warns against those who wish to simplify at all costs, calling them ‘dangerous dogmatists and tedious obscurantists’, and Morin (2008: 57) refers to the ravages caused by simplifying visions, arguing that ‘[m]uch of the suffering of millions of beings resulted from the effects of fragmented and one-dimensional thought.’ In practice, the provisional imperative thus leads to action that ‘enlarges the field of vision, opening up new possibilities and revealing the abundance’ (von Foerster & Poerksen, 2002: 36), and in this way, allows us – at minimal – to remain open to the truths of others, and trust that others will remain open to our truths.


  1. Allen, P.M. 2001. ‘A complex systems approach to learning, adaptive networks’, International Journal of Innovation Management 5(2), 149 – 180.
  2. Allen, P.M. 2000. ‘Knowledge, ignorance, and learning’, Emergence 2(4), 78 – 103.
  3. Badiou, A. 2009. ‘Jacques Derrida’ in Pocket Pantheon, trans. D. Macey. London: Verso, 125 – 144.
  4. Baier, A. 1986. ‘Trust and antitrust’, Ethics 96, 231-260.
  5. Bakhtin, M. 1984. Problems of Dostoevsky’s Poetics, ed. and trans. C. Emerson. Minneapolis: Minnesota University Press.
  6. Burtynsky, E. 2006. Manufactured Landscapes. A film by J. Baichwal, Canada: Mongrel Media.
  7. Cilliers, P. 2010. ‘Difference, identity, and complexity’, Philosophy Today, Spring 2010, 55 – 65.
  8. Cilliers, P. 2005. ‘Complexity, deconstruction and relativism’, Theory, Culture & Society 22(5), 255 – 267.
  9. Cilliers, P., de Villiers, T. & Roodt, V. 2002. ‘The formation of the self. Nietzsche and Complexity’, South African Journal of Philosophy 21(1), 1–18.
  10. Chu, D., Strand, R. & Fjelland, R. 2003. ‘Theories of complexity: common denominators of complex systems’, Wiley Periodicals 8(3), 19 – 30.
  11. Colebrook, C. 2004. Irony: The New Critical Idiom. London: Routledge.
  12. Culler, J. 1983. On Deconstruction: Theory and Criticism after Structuralism. London: Routledge.
  13. Derrida, J. 2005. ‘The principle of hospitality’, trans. A. Thompson, Parallax, 11(1): 6 – 9.
  14. Derrida, J. 2002a. ‘As if it were possible, “within such limits”…’ in Negotiations: Interventions and Interviews, 1971 – 2001, ed. and trans. E. Rottenberg. Stanford: Stanford University Press, 343 – 370.
  15. Derrida, J. 2002b. ‘Ethics and politics today’ in Negotiations: Interventions and Interviews, 1971 – 2001, ed. and trans. E. Rottenberg. Stanford: Stanford University Press, 295 – 314.
  16. Derrida, J. 2002c. ‘The right to philosophy from a cosmopolitical point of view: the example of an international institution’ in Ethics, Institutions, and the Right to Philosophy, ed. and trans. P.P. Trifonas, Maryland: Rowman & Littlefield Publishers,1– 18.
  17. Derrida, J. 2002d. ‘Negotiations’ in Negotiations: Interventions and Interviews, 1971 – 2001, ed. and trans. E. Rottenberg. Stanford: Stanford University Press, 11 – 40.
  18. Derrida, J. 1999. ‘Hospitality, justice and responsibility: a dialogue with Jacques Derrida’ in Questioning Ethics: Contemporary Debates in Philosophy, eds R. Kearney & M. Dooley. London, New York: Routledge, 65 – 83.
  19. Derrida, J. 1988. ‘Afterword’ in Limited Inc., ed. G. Graff, trans. S. Weber. Evanston: Northern Western University Press, 111 – 160.
  20. Derrida, J. 1981. ‘Positions’ in Positions, trans. A. Bass. Chicago: University of Chicago Press, 37 – 96.
  21. Derrida, J. 1979. ‘Living On: Border Lines’, trans. J. Hulbert in Deconstruction Criticism, ed. H.I. Bloom. New York. Continuum, 75 – 175.
  22. Derrida, J. 1978. ‘From restricted to general economy: a Hegelianism without reserve’ in Writing and Difference, trans. A. Bass. Chicago: University of Chicago Press, 251 – 277.
  23. Franzen, J. 2010. Freedom. Toronto: HarperCollins Canada.
  24. Forst, R. 2007. ‘“To tolerate means to insult”: toleration, recognition, and emancipation’ in Recognition and Power, eds. B. Van den Brink & D. Owen. Cambridge: Cambridge University Press, 215 – 237.
  25. Hargrave, T.J. 2009. ‘Moral imagination, collective action, and the achievement of moral outcomes’, Business Ethics Quarterly 19(1), 87 – 104.
  26. Horgan, J. 1995. ‘From complexity to perplexity’, Scientific America, June 1995, 104 – 109.
  27. Kant, I. 1993. Grounding for the Metaphysics of Morals, 3rd ed., trans. J.W. Ellington. Indianapolis: Hackett.
  28. Kearney, R. 1988. The Wake of Imagination: Towards a Postmodern Culture. Minneapolis: University of Minnesota Press.
  29. Levi-Strauss, C. 1966. The Savage Mind. Chicago: University of Chicago Press.
  30. Luhmann, N. 2000. ‘Why does society describe itself as postmodern?’ in Observing Complexity: Systems Theory and Postmodernity, eds. W. Rasch & C. Wolfe. Minneapolis, London: University of Minnesota Press, 35 – 49.
  31. Luntley, M. 2003. ‘Ethics in the face of uncertainty: judgement not rules’, Business Ethics: A European Review 12(4), 325 – 333.
  32. Mandela, N. 1994. Long Walk to Freedom: The Autobiography of Nelson Mandela. London: Little, Brown.
  33. May, T. 1995. The Moral Theory of Poststructuralism. Pennsylvania: Pennsylvania State University Press.
  34. Mensch, J.R. 2003. Ethics and Selfhood. New York: State University of New York Press.
  35. Montuori, A. 2003. ‘The complexity of improvisation and the improvisation of complexity: social science, art and creativity’, Human Relations 56(2), 237 – 255.
  36. Morin, E. 2008. On Complexity, trans. S.M. Kelly. Cresskill: Hampton Press.
  37. Morin, E. 2007. ‘Restricted complexity, general complexity’, trans. C. Gershenson, in Worldviews, Science and Us: Philosophy and Complexity, eds. C. Gershenson, D. Aerts & B. Edmonds. Singapore: World Scientific, 5–29.
  38. Morin, E. 1999. ‘Seven complex lessons in education for the future’, trans. N. PollerUNESCO Publication.
  39. Painter-Morland, M. 2006. Redefining accountability as relational responsiveness’, Journal of Business Ethics 66, 89 – 98.
  40. Plato. 1987. The Republic, 2nd ed., trans. D. Lee. London: Penguin Books.
  41. Preiser, R. & Cilliers, P. 2010. ‘Unpacking the ethics of complexity: concluding reflections’ in Complexity, Difference and Identity, eds. P. Cilliers & R. Preiser. Dordrecht: Springer, 265 – 287.
  42. Rosen, R. 1985. Anticipatory Systems: Philosophical, Mathematical and Methodological Foundations. Oxford: Pergamon Press.
  43. Sontag, S. 2007. ‘At the same time: the novelist and moral reasoning: the Nadine Gordimer lecture’ in At the Same Time: Essays and Speeches, eds. P. Dilonardo & A. Jump. New York: Farrar, Straus and Giroux, 210 – 230.
  44. Verstraeten, J. (ed.) 2000. Business Ethics: Broadening the Perspective. Leuven: Peeters, xii – xiv and 1 – 8.
  45. Von Foerster, H. 1990. ‘Ethics and second-order cybernetics’ Opening address for the international conference, Systems and Family Therapy: Ethics, Epistemology, New Methods, Paris, 4 October 1990.
  46. Von Foerster, H. & Poerksen, B. 2002. Understanding Systems: Conversations on Epistemology and Ethics. New York: Kluwer Academic/Plenum Publishers.
  47. Williams, Z. 2003. ‘The final irony’, The Guardian, Saturday 28 June 2003.
  48. Wimsatt, W.C. 2008. ‘Aggregativity: reductive heuristics for finding emergence’ in Emergence: Contemporary Readings in Philosophy and Science, eds. M.A. Bedau & P. Humphreys. Cambridge, MA: The MIT Press, 99 – 110.
  49. Woermann, M. 2010. ‘Corporate identity, responsibility and the ethics of complexity’ in Complexity, Difference and Identity, eds. P. Cilliers & R. Preiser. Dordrecht: Springer, 167-192.
  50. Wood, D. 1999. ‘The experience of the ethical’ in Questioning Ethics: Contemporary Debates in Philosophy, eds. R. Kearney & M. Dooley. London, New York: Routledge, 105 – 119.


  1. The notion of tolerance is increasingly viewed in a negative light in the extant literature, and is often understood as the passive acceptance of difference, and, hence, denotes an unwillingness to engage in other perspectives. Indeed, Rainer Forst (2007) goes as far as to suggest that tolerance could be translated as a kind of insult to difference, and that ‘recognition’ is a better term for designating the positive qualities traditionally associated with the notion of tolerance. In the context of this paper, the notion of tolerance should be read in a positive light, as both respect for, and an engagement with, human diversity.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.